This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TDA4VM: CPSW9 benchmarks

Part Number: TDA4VM

Team,

Do we have benchmark data about throughput on the integrated Ethernet switch? I couldn't find anything in the Linux SDK doc. And the relevant section in the RTOS doc is empty: 

https://software-dl.ti.com/jacinto7/esd/processor-sdk-rtos-jacinto7/08_02_00_05/exports/docs/ethfw/docs/user_guide/val_notes_mainpage.html#val_perform_measurements

Thanks,
  Robert

  • Hi Robert,

    EthFW Performance data is available in the SDK datasheet. Please follow the instructions in SDK user guide to request access.

    https://software-dl.ti.com/jacinto7/esd/processor-sdk-rtos-jacinto7/08_02_00_05/exports/docs/psdk_rtos/docs/user_guide/datasheet_sdk_j721e.html

    Here is the snapshot from the datasheet.

    1.1.2.4.2. UDP (IP stack) Performance
    1.1.2.4.2.1. CPSW9G - Main domain R5_0 core 0 (mcu2_0)
    • Main domain R5_0 at 1GHz
    • RGMII interface at 1Gbps

    Single Direction Test

    Test Measured Throughput (Mbps) CPU Load (%)
    UDP RX 168.0 93
    1.1.2.4.2.2. CPSW2G - MCU domain R5 core 0 (mcu1_0)
    • MCU domain R5_0 at 1GHz
    • RGMII interface at 1Gbps

    Single Direction Test

    Test Measured Throughput (Mbps) CPU Load (%)
    UDP RX 142.0 95

    Note:

    1. TCP/IP throughput results are measured using Enet LLD lwIP example application built with the following modifications which are not PDK’s default settings:
      • lwIP example app built with performance optimizations (OPTIMIZATION=PERFORMANCE passed to make command)
      • Enable Thumb2 mode (CFLAGS_INTERNAL += -mthumb)
    2. Current performance numbers are preliminary as throughput profiling is not done in optimized environment (i.e. putting frequently used functions in fast memory, use of pacing using ring monitor, tuning of descriptors, etc.)

    Regards,
    Stanley

  • Hi Karthik,

    We had knowledge of those but we were seeking for more detailed benchmarks.

    Are performance numbers available for the MAIN R5Fs, A72s and C71x from the CPSW0 switch 10G port in a scenario using the built in ALE for frame recognition and the subsequent transfer to the destination core ? We would also be interested in latency values on top of transfer rates and Core usage in this scenario.

    Is it possible to perform port mirroring of all incoming and outgoing traffic without CPU intervention within the CPSW0 switch ?

    Best regards